Report on the INEX 2006 Heterogeneous Collection Track

  • Citation-Key:
    Frommholz/Larson:07
  • Title:
    Report on the INEX 2006 Heterogeneous Collection Track
  • Author(s):
    Ingo Frommholz
    Ray Larson
  • Journal:
    SIGIR Forum
  • Volume:
    41
  • Number:
    1
  • Year:
    2007

Abstract:


XML collections that consist of subcollections from different sources pose a challenge with respect to syntactic, semantic and genre heterogeneity, because they are based on different DTDs or schemas, cover various topics and consist of diverse document types. This led to the establishment of the Heterogeneous Track at INEX 2006. The goal of the track was to set up a testbed consisting of several different and diverse collections and defining retrieval tasks and appropriate topics. These are the foundations for the Heterogeneous Track at INEX 2007, where the focus is on run submissions, relevance assessments and proper evaluation of the proposed methods dealing with a heterogeneous collection.

Fulltext as PDF